From Retrograde Menstruation to Endometriosis – A Possible Pathophysiological Role of L1, a Cell Adhesion and Migration Protein

Introduction: The L1 transmembrane glycoprotein regulating cell migration and adhesion is known to play an important role for the development of the nervous system. It is also present at high levels in several inflammatory and malignant disorders. L1 is highly expressed in uterine and ovarian carcinomas. High amounts of soluble L1 in blood samples of endometrial cancer patients are a marker for early recurrence and poor prognosis. The expression and cleavage of L1 seems to promote dissemination of tumor cells by enabling cell migration and adhesion. The cell-surface expressed L1 of tumor cells can be transformed into soluble L1 by the matrix metalloproteinase ADAM10 related cleavage of its ectodomain. Endometriosis is a benign and mostly progressive disease with a high prevalence in the female population. Nevertheless it shows particularly behaviour of malignant diseases like deep invasion in different tissues. This behaviour is based on migrating endometrioid tissues build by epithelium and stromal cells of modified endometrium, adhering to distant tissues. We hypothesize that L1 could be involved in dissemination of endometriotic cells after retrograde menstruation, possibly promoted by the acting of ADAM10. Material and Methods: Endometrial and endometrioid tissue as well as serum samples were obtained from patients undergoing laparoscopy after providing written informed consent under a study protocol. We investigated 33 patients in the endometriosis group (Eo) and 18 controls (Co). Immunohistochemistry was performed on serial 4µm tissue sections from endometriotic lesions and endometrium of women with and without endometriosis. Cell cultures were prepared from endometrium and endometriosis implants. Total RNA was extracted from tissue and cell cultures (endometrial and endometriosis stromal and epithelial cells). Real-time PCR was performed with specific oligonucleotide primers which were designed to amplify sequences from human L1 and ADAM10 mRNA, and CoxL served as housekeeping gene. To detect soluble L1 in serum samples of patients with and without endometriosis we used a specific ELISA. Results: Immunohistochemical staining showed expression of the L1 adhesion molecule preferentially in epithelial sections, but also some weak staining in the stromal compartment. L1 was more expressed in endometriosis tissue samples than in endometrium of healthy controls. Endometrium from patients with endometriosis showed more intense staining of L1 compared to endometrium from healthy controls. Real-time PCR demonstrated higher L1 expression in endometriotic tissue than in controls. In RNA from cell cultures of endometrium, L1 was higher expressed in epithelial (Eo 66% vs. Co 10%), than in stromal cells (Eo 50% vs. Co 8%). Moreover, ADAM10 expression shows similar distribution in epithelial (Eo 100% vs. Co 40%) and stromal cells (Eo 99% vs. Co 37%). In serum samples of women with and without endometriosis however, we could not detect statistically significant differences in soluble L1. Conclusion: L1 protein was most prominent in endometriotic tissue, followed by endometrium of endometriosis patients. The weakest L1 amount could be seen in the endometrium of healthy controls. We demonstrated a higher expression of L1 mRNA in epithelial cells compared to stromal cells. Similar to L1 mRNA distribution, ADAM10 mRNA was higher expressed in endometriotic tissue than in healthy control endometrium. We could detect soluble L1 resulting from ectodomain cleavage in blood samples. Unfortunately, there was no significant difference in soluble L1 in serum samples from endometriosis patients compared to controls detectable. So, in contrast to patients with endometrial cancer, L1 in blood samples of endometriosis patients can not serve as a marker or prognosis factor for endometriosis. Further investigation are needed to show, if it is possible to inhibit the action of L1 and prevent or treat endometriosis.